Interstitial cystitis (IC) is a severe debilitating bladder disease of unknown etiology and no cure. A recent double-blind trial using intravesical Bacillus Calmette Guerin (BCG) to treat IC demonstrated a 60% clinical response rate to a single six week course of BCG. When subjects who responded to BCG were followed for a minimum of two years, 89% continued to have an excellent response, despite no additional treatment of their IC. The durability of this treatment leads one to speculate on the mechanism in which intravesical BCG may treat interstitial cystitis. There is evidence that interstitial cystitis may be mediated by a T-Helper Cell type-2 (Th-2) response within the bladder. Cytokine analysis from the urine of IC subjects showed elevated levels of Interleukin-6 and inhibitors of interleukin-2, suggesting a Th-2 response. In addition, similar autoantibodies have been identified in both atopic dermatitis, a Th-2 mediated disease, and interstitial cystitis. However, the role of the immune system in the etiology of IC remains controversial. We hypothesize that interstitial cystitis is a Th-2 mediated disease leading to chronic inflammation and that intravesical BCG is effective by converting the cytokine milieu to a Th-1 profile, leading to reparative conditions and long-term clinical response. Specifically, this study will: 1) determine the urine cytokine profiles in subjects meeting the NIDDK criteria for interstitial cystitis and in health control subjects; 2) determine in a blinded fashion the changes in urinary cytokines during six weekly instillations of either bacillus Calmette-Guerin (BCG) or placebo and at regular intervals during a 6 month follow-up; 3) correlate changes in cytokines with clinical response; and 4) determine whether a certain cytokine profile cytokine profile can predict clinical response to intravesical BCG therapy. This study will involve subjects enrolled in our present clinical trial of intravesical BCG therapy for IC. Cytokine levels will be determined in triplicate by enzyme-linked immunosorbant assays and normalized against urine creatine. Study results will be analyzed by non-parametric methods. In addition, a receiving operating characteristic analysis will be completed to determine the critical cytokines levels for predicting clinical response to treatment. In summary, this study will determine the cytokine profile in IC subjects and healthy subjects. By correlating the changes in cytokine levels before, during and following intravesical BCG therapy, we will establish the role of these cytokines in IC and use the pattern of change as a means to predict subject response to therapy. Additionally, this study will open a new avenue of research with specific long-term potential for the development of more effective, less toxic treatments of IC.